A new method for detecting i2c/SPI vulnerabilities in IOT devices based on Ai Neuro Symbolic framework.

A cybersecurity IoT system for black box testing simulates real-world attacks on IoT devices with no internal knowledge, focusing on identifying vulnerabilities in device hardware, software, and network communication by mimicking external attackers, using tools for scanning (ports, services), firmware analysis (firmware extraction, reverse engineering), and network sniffing to uncover weaknesses like default credentials, insecure protocols, and data leaks, ultimately building trust and improving security posture.

IOT devices are ubiquitous in our daily lives – whether at home with connected home automation devices or at work with connected factories, hospitals and even connected cars. According to data collection and visualization firm Statista, there were an estimated 15.9 billion IoT devices in use in 2023, and that number is expected to grow to more than 32.1 billion by 2030. As enterprises globally have transformed their processes over the past decade with more embedded, IOT-powered intelligence, these billions of connected devices have also become a target for cybercriminals.

IOT attacks in the EU are rampant, focusing on DDoS, ransomware, and exploiting weak device security for broader network access, with recent threats like the Kimwolf botnet targeting Android IOT devices and the Clickfix malware hitting hotels via social engineering, prompting EU action like the Cyber Resilience Act (CRA) to mandate better security from the start, enforcing secure-by-default design, updates, and clear vulnerability reporting for manufacturers. We also need to take cybersecurity technical debt into the research project and help to improve or remove old devices with better and newer devices. We will address this problem during duration of the project with investigation of the problem and present a better solution that it is possibly today.

We use engineering research methods:
Quantitative Methods: Use of data, statistics, measurements and numerical models to understand and predict phenomena.
Qualitative Methods: In-depth analysis of systems and processes.
Experimental Design: Systematic setup of experiments to test hypotheses.
Modeling and Simulation: Creating abstract representations of complex systems to predict behavior.
Technical/Practical Methods: Focus on developing concrete solutions, prototypes and implementations.

The developed system can help companies for external testing and validation of their IOT/OT devices in relation to NIS2 and thus improve their products and provide documentation in relation to NIS2. In the long term, the system can also help companies with security within the OT area better than today.